The long-term goal of this project is to examine lexical segmentation in aphasia, and to use the resulting data to help understand how humans accomplish the difficult task of recognizing individual words in connected speech. The strategy behind the research is to identify patients with specific impairments in targeted aspects of language processing including lexical access, and the discrimination of putative acoustic-phonetic word boundary cues, and examine how they interpret speech sequences known as oronyms in which positing different word boundaries leads to recognizing different words (e.g., kidnap/kid nap). It will address four issues: (1) whether lexical segmentation is the results of a discrete process, or a byproduct of lexical access, (2) how the acoustic form of word onsets affects lexical segmentation, (3) what the timecourse of segmentation disambiguation is, and (4) what factors modulate interword competition in lexical access and/or segmentation. These issues will be examined through a series of offline discrimination tasks and online paradigms including cross-modal lexical priming and word monitoring that provide implicit measures of aphasic and unimpaired listeners' interpretation of oronyms. This research will provide both individual and group studies of lexical access and segmentation and their impairment. At present there are no published studies examining lexical segmentation in aphasia. As segmentation is one of the central problems of spoken word recognition in connected speech processing, this work addresses a critical gap in our understanding of aphasic disturbances of spoken language comprehension. In addition to characterizing the nature of segmentation processes in aphasia, this work will provide a new source of converging evidence to understand the organization of spoken word recognition processes in normal listeners. It is expected that the understanding gained will ultimately be useful to clinicians, therapists and theorists.